Lesson 9 – Vorticity and Potential Vorticity Conservation of vorticity – Many motions in the atmosphere and ocean are rotational in nature. Hurricanes and tornados are two obvious examples but more subtle examples also exist such as the presence of vorticity in a turbulent medium and aerodynamic lift. It is of interest for us to derive and examine the conservation of the rotational characteristics of the fluid system as a way to gain further physical insight into the physical nature of the problem of interest. We learned last semester that the local vorticity, u , of a fluid is equal to twice the value of an equivalent solid body rotation of the same fluid element. Given that vorticity is related to the velocity field, we can utilize the equations of motion in a rotating frame as derived in chapter 6 as a starting point for obtaining the equation for conservation of vorticity and, consequently, also a quantitative means of predicting the rotational character of the fluid. In chapter 3 we derived the equations of motion assuming a f-plane approximation and small variations in density. The results are given by Du 1 p fv 2 u Dt x Dv 1 p fu 2 v Dt y Dw 1 p g 2u Dt z u 0 (1) We will also assume that the horizontal flow is 2-D so u ux, y and v vx, y in equation (1). In vector form, the first three equations above can be expressed as: u 1 u u f u p g 2u t ^ (2) ^ Where f fo y k and g g k and we have expanded out the material derivative into its two components. We are interested in taking the curl of equation (2) to convert the velocity vectors in the above equations to vorticity vectors. To make this process easier, we convert the non-linear advection term to a form much more tractable for our calculations: 1 1 u u u u u u u 2 u 2 2 Equation (2) is then expressed as: u 1 1 u 2 u f u p g 2u t 2 Taking the curl of equation (3) and using the vector identity: 1 (3) A B B A A B A B B A We obtain a general form of the conservation of vorticity equation: D 1 f u v 2 p 2 Dt (4) In the above equation, we assumed that all body forces were eliminated since they were assumed to be conservative so g . Also, notice the third term is due to the -plane approximation, u f v . Let us examine the physical characteristics of each term: D u - This is the material change of vorticity. It is a measure of the Dt t time rate of change of vorticity in a fluid element as that element travels in the flow field. 1. 2. f u - The first term, u , is measure of the stretching and tilting of vortex lines in the fluid. A vortex line is just a line of constant vorticity that is parallel to the vorticity vector field in the fluid domain. Just like streamlines, the intensity of the vorticity is represented by the density of vortex lines in the fluid domain. Thus, if these lines are stretched in a region, the material change in vorticity is positive. This is the same as the “ballerina effect” concept in angular momentum. For a given vortex tube, if we stretch the tube so that it has a smaller radius and greater height, the density of the vortex lines increase and the overall vorticity or rotation will increase accordingly as shown in figure 1 Vortex stretching Figure 1 – “Stretching” of a vortex tube indicating a greater density of vortex lines in the right vortex tube and therefore a greater vorticity and rotation for the right tube. The second term f u , is a measure of the stretching of the streamlines of the fluid in the direction of the Coriolis vector (usually the vertical direction). If the streamlines are stretched in the vertical direction, then there is a material increase in 2 D in the Dt conservation of vorticity equation indicates that planetary rotation leads to non-zero vorticity in the fluid domain. vorticity in that region. The fact that the term f u is related to 3. 1 2 p - This is a measure of the baroclinic contribution to the material change of vorticity. To see this, assume that the fluid is barotropic so that p p then dp 1 1 1 p 2 and then 2 p 2 2 0 . This shows that d c c barotropic fluids make no contribution to the material change in vorticity. In other words, the baroclinic properties of a fluid provide a contribution to the net increase or decrease of vorticity of the fluid. 4. 2 - This is a measure of the diffusion or “spreading out” of the vorticity due to frictional properties of the fluid medium. 5. v - This terms shows that vorticity will increase due to the presence of a southerly component to the wind field and variations in latitude. Potential vorticity - preliminaries: All of the physical systems we have examined thus far in this course have assumed the bottom layer as flat in all directions. In order to consider the orographic or bathymetric variations of the fluid boundary, we need to develop a physical quantity called potential vorticity and examine its conservation equations. For starters, let us consider the vertical component of equation (4) where the vertical vorticity component is ^ . Recall the Coriolis vector is in the k direction in the -plane approximation as shown in last equation in (1). Finally assume the fluid is incompressible so 0 and u 0 . Equation (4) can then be simplified to: D w f v Dt z Notice that the Coriolis vector has no time dependence and depends spatially on y so we can represent the material derivative of the Coriolis vector as v . In other words, Df f u f v . This is just the third term in the above equation so we can rewrite Dt t the above expression in the simple form. D f f w Dt z (5) 3 The term f is called the absolute vorticity since it is the sum of the vertical component of the relative vorticity, , as well as the effects of the earth’s rotation, f. Equation (5) physically represents why divergent high pressure systems rotate anti-cyclonically and convergent low pressure systems rotate cyclonically. From the incompressibility condition, the vertical velocity variations are related to the horizontal u v w divergence by h u . Equation 5 is then represented as: z x y D (6) f f h u Dt The above equation shows us that when there is a net outflow (horizontal divergence is positive) then there is a material decrease in the absolute vorticity. Now consider a high pressure system where the initial relative vorticity is 0 and planetary rotation remains constant. Since high pressure systems have a net outflow at the surface, there will be material decrease in absolute vorticity leading to the formation of negative relative vorticity (clockwise or anti-cyclonic rotation). The reverse occurs in the case of low pressure system where there is a net inflow leading to the development of positive relative vorticity (counterclockwise or cyclonic motion). Shallow water system revisited: Let us consider a shallow water system like in the last chapter but this time we will assume minor variations along the ocean bottom as shown in figure 2. h H z=0 Ocean bottom Figure 1 – Vertical profile of the ocean domain showing small surface displacement superimposed over small displacements in the ocean bottom as compared to a constant reference height H. In this case, the net ocean depth, h, is the sum of the surface displacement, 2 , displacements in the ocean bottom, 1 and a constant reference height, H . We can combine the surface and bottom displacements into one small displacement term 1 2 so the ocean height can be expressed as hx, y H x, y . Integrating the continuity equation from z=0 to z=h, we obtain1: 1 Recall we assumed in the beginning of this chapter that the horizontal motion was 2-D. 4 u v u v wh w0 dz h . x y x y 0 Now, clearly the reference position does not change so w0 0 . Further, the vertical velocity is equal to minus the material changes in the ocean surface and bottom so h wh u v Dh h Dt x y The above equation can be rearranged to the following form: u v 1 Dh h u h Dt x y (7) Conservation of potential vorticity: We can substitute the horizontal divergence in equation (7) into equation (6) to obtain the equation: D f f 1 Dh 0 Dt h Dt Multiplying both sides of the above equation by 1 and observing that h D 1 1 Dh , we obtain the equation of conservation of potential vorticity: 2 Dt h h Dt D f 0 Dt h (8) f We define the quantity in equation (8) as the potential vorticity. h Physically equation (8) shows us that, for an inviscid and incompressible fluid2, the potential vorticity of a fluid parcel is conserved along its motion in the atmosphere or ocean. This is an important and simple result in Geophysical fluid dynamics. Another consequence of equation (8) is the westward intensification of an ocean basin seen in the previous chapter provided we ignore the friction-free assumption leading to equation (8). We can imagine a parcel with zero initial vorticity following the ocean currents around the Atlantic Ocean basin. As the parcel travels northward along the western boundary of the basin (US east coast), the planetary vorticity will increase; requiring a new decrease in relative vorticity. The wind stress field also has a negative vorticity that increases with latitude. The superposition of these negative vorticities must be 2 The condition of incompressibility in the atmosphere might seem too restrictive. We can instead assume a barotropic fluid where the Boussinesq approximation holds. 5 compensated for by an increase in the vorticity due to frictional shears at the ocean surface. Since friction from eddy currents is proportional to the mean velocity of the flow field, the northward moving current must increase. The opposite effect occurs along the eastern boundary of the ocean basin. The southward moving current must decrease to counter the increase in absolute vorticity so that currents are slower along the eastern boundary and intensified along the western boundary. Kelvin’s circulation Theorem*: Returning to the general conservation of vorticity equation D 1 f u v 2 p 2 Dt o The above equation has assumed conservative body forces. If we further assume a barotropic and inviscid fluid, the vorticity equation becomes: D f Dt f u (9) and we can find that, provided that density variations are small and we can use the Boussinesq approximation, the circulation of the absolute vorticity is conserved. Da 0 where a f dA Dt A (10) This conservation of circulation is Kelvin’s Circulation theorem for a fluid in a rotating frame. The theorem shows us that the effects of friction, non-conservative body forces and baroclinicity lead to the development or deterioration of vorticity in the observed physical system. To derive equation (10), apply Gauss theorem to a , D D f dA f dV Dt A Dt V In applying the material derivative to the above equation, we cannot assume that the volume is fixed in space since we are interested in the spatial variations in the vorticity and not just the time dependence. Using Reynolds transport theorem3 along with the Boussinesq approximation, we can take the material derivative of the integral and obtain: 3 Reynold’s Transport Theorem allows us to consider time derivatives over a volumetric integral where the boundaries can vary. It is expressed generally as: D D dV u dV Dt V Dt V 6 D D f dV f dV Dt V Dt V (11) We now need to swap the divergence term with the material derivative, we notice that for a general vector, A D D A A A u Dt Dt Or in index notation: D D Ai Ai Dt xi xi Dt xi A j ui x j Swapping the material derivative with the divergence in equation (11) then leads to the result: D a D D f dV f f u dV Dt Dt Dt V V But the conservation of vorticity as expressed in equation (9) shows us that the argument for the above integral is equal to 0 and Da 0 Dt thus proving Kelvin’s circulation Theorem. To prove this theorem, one must apply a change of coordinates between a fixed and moving frame and apply the Jacobian. We are interested in applying Reynolds transport Theorem where the Boussinesq D D dV dV Dt V Dt V approximation also holds so we will use the relation: 7